DESCRIPTION: The overall goal of this project is to test the hypothesis that stable translocations determined in single chromosomes using fluorescence in situ hybridization (FISH) can be used as a rapid, intermediate endpoint to assess the efficacy of a potentiator of tumor response to fractionated irradiation in the clinic. The applicants believe that this will have significant practical value by allowing the assessment of any new modifying agent (such as a radiation sensitizer of hypoxi cytotoxin) without having to perform a major clinical trial with radiation using local control as the endpoint with he inherent cost and time taken to achieve such a result. There are currently a large number of agents that are being considered for clinical trials with radiation as potential tumor modifiers (i.e., tirapazamine, carbogen with nicotinamide, recombinant human erythropoietin, fluorocarbon emulsions, such as perflubron and others ). It will be extremely difficult to test all of these using clinical outcomes as the endpoint, and for many agents, there are such questions as the need to give the agent with every dose fraction and the doses that should be used. If the assay that is proposied can be used to assess the likely efficacy of these modifiers, it will, they believe, be a significant advance for radiation therapy. The assay that is proposed, the assessment of stable translocations following irradiation, is based on the fact that one of the important lesions killing cells following ionizing radiation is the production of dicentrics with acentric fragments. In addition to this lesion, however, nonlethal stable translocations are also produced, usually in equal numbers to the dicentrics. Any modification of the cell sensitivity to radiation would be expected to equally modify both the incidence of dicentrics and the incidence of stable translocations and, therefore, they propose that the latter can be used to estimate the extent of this modification. They propose to investigate all of the underlying assumptions and hypothesis inherent in this overall application using four different human tumor cells lines growing as xenografts in SCID mice (HT1080, FaDu, SCC61, and A549). They also plan to investigate the uniformity and stability of the karyotype of human tumors split into several parts and transplanted into different SCID mice.